Some Recent Results of the University of Pennsylvania Excavations at Nippur, Especially of the Temple Hill 9781463219987

Citation preview

Some Recent Results of the University of Pennsylvania Excavations at Nippur, Especially of the Temple Hill

A n a l e c t a Gorgiana

256 Series Editor George Kiraz

Analecta Gorgiana is a collection of long essays and

short

monographs which are consistently cited by modern scholars but previously difficult to find because of their original appearance in obscure publications. Carefully selected by a team of scholars based on their relevance to modern scholarship, these essays can now be fully utili2ed by scholars and proudly owned by libraries.

Some Recent Results of the University of Pennsylvania Excavations at Nippur, Especially of the Temple Hill

John Peters

gorgia* press 2009

Gorgias Press LLC, 180 Centennial Ave., Piscataway, NJ, 08854, USA www.gorgiaspress.com Copyright © 2009 by Gorgias Press LLC Originally published in All rights reserved under International and Pan-American Copyright Conventions. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise without the prior written permission of Gorgias Press LLC. 2009

1

ISBN 978-1-60724-485-1

ISSN 1935-6854

Extract from The American Journal of'Archaeology and of the History of the Fine Arts, vol. 10 (1895).

Printed in the LTnited States of America

SOME R E C E N T R E S U L T S OF T H E U N I V E R S I T Y OP PENNSYLVANIA EXCAVATIONS AT NIPPUR, E S P E C I A L L Y OF T H E T E M P L E HELL. [PLATES I I I , I V ,

V.]

The expedition to Babylonia which was sent out under my direction commenced excavations at Nippur February 6th, 1889, and continued them for two months. "Work was resumed January 14th, 1890, and continued for the space of four months. During the first year we worked with a maximum force of 200 men. Naturally a large part of our work was tentative, especially the work upon the temple. During the second year we worked with a maximum force of about 400 m e n ; and while we employed a considerable number of these in digging for tablets and making soundings in various parts of the extensive mounds, by far the larger part of the force was, from the outset, concentrated on the systematic exploration of the Temple Hill. In 1892 a second expedition was sent out under the direction of Mr. J . H. Haynes, a member of my staff during the first two years, who had also had experience with other expeditions, especially the "Wolfe expedition and the expedition to Assos. Mr. Haynes commenced excavations April 11th, 1893, and continued them until April 3d, 1894, resuming work again J u n e 4th, 1894, and continuing down to the present time. His average force of workmen employed has been about 50, and from September 1st, 1893, until November 24th, 1894, this force was concentrated upon the Temple Hill, continuing the systematic excavations which had been carried on at that point, and especially upon the ziggurat of the temple and its immediate surroundings. A considerable section of the temple in front of the ziggurat to the southeast, has been removed, stratum after stratum, and the ziggurat itself followed down from its latest form to its earliest 13

14

JOHN F.

PETERS.

with great care. The complete results of this work cannot be published until Mr. Haynes' expedition has returned, and there has been an opportunity for him and others to work up those results ; but combining what was done under my direction during the first two years with what Mr. Haynes has so far reported for the subsequent two years, I am able to present some sort of a sketch of the history of the great temple of Bel at Nippur. The accompanying plan (Plate in) will give some idea of the appearance of the temple enclosure at the close of the first two years' excavations. The shaded portions on this map represent actual excavations. It will be observed that the ziggurat has wing-like or buttress-like projections on all four sides and is curiously irregular in form. The ziggurat, as here represented, is composed of two stages. About it, on all sides, we find rooms or corridors. The ziggurat, with the various rooms, corridors and the like which surrounded it, was enclosed by a huge wall, which towards the southeast stood to the height of over 60 feet, and was almost 50 feet thick at its base. On the top of this wall, on the southeastern side, we found a series of rooms. There were irregular, tower-like masses at three of the corners of the great wall. The western corner, and a part of the southwestern side near the western corner, could not be found at all, having been apparently destroyed by water. At the eastern corner there was a singular mistake, owing to the lack of instruments of precision, by which the angle was made obtuse instead of right, thus giving the enclosure a curiously irregular shape. The corners of the great enclosure and of the ziggurat itself were not accurately orientated —the northern corner of the ziggurat pointing 12 degrees east of the magnetic north. But to commence with the section A, B : the apparent level of the Shatt-en-Ml, or canal bed, which passes through the mounds of Xippur, is several meters above the true level of virgin soil, the old canal having filled up gradually with washings from the mounds about it. I sank a shaft to the depth of 19.3 metres from the 14 metre level, or 5.3 meters below the canal level, on the outer surface of the great wall, without reaching bed earth; and Mr. Haynes reports a depth of 68£ feet, or almost 21 metres, from the surface of the 14 metre level to virgin clay in a shaft

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

15

sunk by him in the great trench on the temple plateau, which would m a k e virgin soil 7 metres below the canal level. I t h i n k this must have been an exceptional place, and that the true soil level is in general not much more than 5 metres below the canal level. My excavations were actually continued some two weeks after this section was completed, with the result, so far as depths go, of reaching 19.3 metres in the trench outside of the great wall instead of 18, as there indicated; of removing t h e inner wall, and of carrying the trench inside of the great wall and between it and the ziggurat down to 11.5 metres below the 14 metre level, at which depth we found the foundations of walls of the famous Sargon, k i n g of A k k a d e , 3800 B. C. The trenches were connected with one another by a tunnel passing t h r o u g h the great wall, while a similar tunnel led from the trench outside of the great wall, t h r o u g h the mass of rubble before it, to the excavations marked " Shrine of B u r Sin." T h e greatest depth reached under the ziggurat itself is n o t shown by this section, since it was off the line. By means of t h e trench, m a r k e d 56 on the plan, and the tunnel, m a r k e d 10, a point 2 metres below plain level was reached under the western corner of the ziggurat. Commencing at A , on the level of the Shatt-enJSTil, there was a low wall-like line of mounds rising 3 to 4 metres above plain level, with a gate-like breach in the middle. This proved to contain a row of booths or small rooms with walls of mud brick. I n the particular room or booth t h r o u g h which the section line passes, in the northern corner of the room, was found a large number of inscribed objects of ivory, glass, turquoise, agate, malachite, lapis lazuli, magnesite, feldspar, etc., some in the process of manufacture and some complete, together with material not yet worked. All these had been contained in a box which had been buried by the falling in of the earth of the walls or roof and decayed away, leaving signs of its existence in long copper nails, in the position of the objects when found, and in some slight traces of oxidation left in the earth by decaying wood. These objects were found from 1.5 to 2.5 metres below the surface. T h e inscribed objects belonged to kings of Babylon of the Cosssean dynasty, f r o m Burna-Buriash, 1342 B. c., to Kadashman-

16

JOHN P.

PETERS.

Turgu, 1241 B. c. No later buildings had been erected at this point. The walls of the booths rested upon a foundation of earth heaped up for the purpose, but where we carried excavations lower, as in the breach-like opening, we found that other buildings had existed there at an earlier period. Behind this outer low wall-line was a depression of the surface, beyond which again the ground sloped up quite steeply to the summit of the temple plateau, which was 14 metres above plain level. On this slope, at the point indicated in the section, was found the shrine of Amar-Sin or Bur-Sin, as the name is variously read, indicated on the plan at No. 11. This shrine stood on a platform of burned brick. Its walls were built of burned brick laid in bitumen, and from 7 to 14 courses were still in place. Almost all of the bricks bore a brief dedicatory inscription to Bel by Bur-Sin, king of Ur. about 2400 b. c., and longer inscriptions by the same monarch were found on two fine diorite door sockets in the two doorways. This building faced towards the ziggurat, as shown on the plan. Behind it, and belonging to it, was a well, also of burned brick. Apparently there had been statuary and ornamental basreliefs in connection with this shrine. A n excavation of the gully beneath it revealed a pair of clasped hands from a diorite statue, which must have been similar to those found at Tello, and several inscribed fragments, including three fragments of basreliefs. A n archaic looking mortar of volcanic stone was also found at the same place. These objects are reproduced in No. 2 of Plate v. As we found it, this little building faced against a huge towering wall, under the debris from which it had been buried, but at the time of its erection either the wall did not rise above the level of the platform of this temple, or, if it did, there was in it a large opening serving as an entrance to the temple at this point. Walls of brick of Ur-Gur (2800 B. c.) are buried in the great wall. These walls were part of a causeway ascending from a point about on a level with and nearly in front of this shrine of Bur-Sin, to the top of the first terrace of the ziggurat. This shrine, therefore, reminds me by its position of the " high places of the gates," mentioned in the books of Kings (as, for example, n K., 23:8).

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

17

At a later date a certain King Gande scratched his inscription on the side of one of the door sockets of Bur-Sin. Of this king we know nothing, except that he caused his name to be scratched on the work of several older kings at Nïppur; and his inscriptions also stand by themselves on three large, rude marble stones. His inscriptions are extremely barbarous in appearance. On top of the ruins of this shrine of Bur-Sin we found a poor wall of mud brick, with no clue to its age, and above this a mass of débris which had fallen from the great wall above. The great wall was of really colossal proportions. It had a slope of one in four. At the bottom it was 15 metres in thickness, and at the top, as it at present stands, 9 metres. For 14 metres below the level of the plateau this wall was built entirely of unbaked brick, but below this, for 5.3 metres, it consisted of earth faced with a casing of baked brick .90 m. in thickness, and the slope of this lower part was loss than that of the upper. That the wall was not homogeneous and all constructed at one time was clear, among other things, from the fact that a portion of the brick causeway, by which in Ur-Grur's day access was had to the upper stages of the ziggurat, was imbedded in it. A wall had evidently existed at this point from time immemorial, repaired and built upon by men of many ages, until it reached its present height. Originally, as shown by the fragments of a transverse wall found at a low level, there was an entrance over this wall on the southeastern side, by means of steps or an inclined plane; and as late as the time of Bur-Sin there was still an entrance at this point. A t the time of the last great reconstruction this wall was raised to a much greater height, perhaps for purposes of defence, and there is no trace of an entrance in front. Rooms were built upon this last wall, as shown in the plan. On its inner side this wall was intended, in the last reconstruction, to be above ground to the depth of 5.5 metres only, the rest being a retaining wall to enclose a terrace. That terrace, as we found it, was composed largely of débris, but in many places, especially along the line of the walls, was found a filling of unbaked bricks in large square blocks. A somewhat similar wall surrounded the temple enclosure on all sides, and each side has roughly an outside measurement of 200 metres, excepting for the difference caused by th&

18

JOHN P.

PETERS.

fact that at the eastern corner, as already stated, by some mistake the angle was made obtuse instead of right. The corners of the wall, as of the building in general, point approximately toward the cardinal points, in such a manner, however, that the northern corner is 12° east of north. It may be added that I observed similar inaccuracies of orientation at Mughair (Ur) and W a r k a (Erech). The orientation of Babylonian buildings was merely approximate, and I am inclined to think that it was determined originally by the trend of the valley and the prevailing winds rather than by astronomical observations. Within this outer wall on the southeastern side of the temple there was, as will be seen from both the section and the plan, an inner wall with two almost circular towers. The depth of this wall was 9.5 metres. It evidently was in existence at the time of Ur-Grur, and was perhaps first built by him. It must have been rebuilt and added to from time to time until it received its present form at the time of the last great reconstruction. As we found it, in the upper 5 metres of its surface it was beautifully plastered or stuccoed, while the lower 4.5 metres consisted of plain unplastered blocks of unbaked brick. A section of this wall was ultimately removed as far as the tower marked 63 in the plan, in trench 1; and that trench itself was carried a metre lower throughout than is shown in the section. This was done by me. Mr. Haynes more than doubled the breadth of this great trench, extending it toward the northeast, and also carrying it northwestward through the projecting southeastern wing of the ziggurat up to the line of the inner and more ancient ziggurat. He also removed all additions to the ziggurat itself until he had reached the original structure of Ur-Gur. As will be seen from the plan and section, the great trench was carried in the first two years only up to the southeast wing of the ziggurat; but another trench was carried around the entire ziggurat, and that structure, a solid mass of brick and mud-brick, was explored through all its strata by means of tunnels and cuts. By this means we were able to ascertain that there was another and older ziggurat inside of that which our excavations had laid bare. The cut through the core of the ziggurat (No. 52) showed us that the depth of the mass of unbaked brick was 23 metres

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

19

from its highest point. "With the exception of the outer casing belonging to the last great restoration, this mass was homogeneous in construction, consisting of unbaked bricks of small size, and in shape not unlike the ordinary bricks in use to-day. These were the bricks of Ur-Grur. They were laid in three different ways: first a layer on the edge, with the flat sides out; then a layer on the edge, with the ends out; and then a layer on the flat side, with the edges out. These bricks were often somewhat crushed out of shape by the weight resting upon them. Below the ziggurat, at the point of cutting (No. 52), we found first a metre of black ashes, and then a metre of earth, with occasional fragments of pottery. "When we tunneled beneath the ziggurat at its western corner, 26 metres below the level of the summit, we found a drain of pottery rings, a fragment of a wall of baked bricks, plano-convex, the convex surface marked with thumb grooves, precisely similar to those shown on figures 15 and 16; an illegible fragment of an unbaked tablet, and a beautiful jade axe-head. In a cutting at the other end of the ziggurat (No. 53), which descended about 9 metres from the top, we found so many bricks and the like as to suggest the existence on the summit of the ziggurat at some period, of a brick structure of some sort; but all surface layers of the ziggurat of the later and earlier periods alike were so ruined and worn away by the action of water that it was impossible to reach certainty on this matter. As will be seen from the plan, the ziggurat, as discovered, was peculiarly irregular in structure. On both sides of the northern corner (No. 54) was a paneled wall of brick. This is a part of the ziggurat of Meli-Shiha. Everywhere else his ziggurat was buried under a new wall of huge blocks of mud-brick. Dotted lines at Eos. 38, 7, 49, 8, etc., on the plan show remnants of older brick structures in various parts of the temple area, notably a grand tower at the northern corner (No. 38), buried in the mudbrick buildings of the last restoration. Space will not permit an extended description of the meaning of the accompanying plan. Excavations were made, as will be seen, over a large extent of space, and in a general way the temple in its last reconstruction was laid bare.

20

JOHN

P.

PETERS.

In view of the great size ot the temple—which covered within its inner walls a surface of about eight acres—it was impossible to excavate the whole of it systematically, removing stratum after stratum. For this purpose we chose, as shown in Fig. 2 a section immediately in front of the ziggurat to the southeast, between the ziggurat and the great wall, and conducted a large trench with a view of determining the successive strata. This enabled us to treat also the wall on one side and the ziggurat on the other. Ultimately it was found necessary to remove consid-

F I G . 2.—EXCAVATIONS ON THE S . E . S I D E OF THE ZIGGURAT OF THK T E M P L E OF B K L ( J u n e 5, 1894).

erable portions of the ziggurat in order to get at the original constructions, and to carry the large trench around the ziggurat on all sides. Wells and similar shafts were sunk at other parts of the temple, wherever a favorable opportunity seemed to present itself, for the purpose of confirming, checking and reinforcing the results obtained from the excavation of the space in front of the ziggurat; but in giving the result of the excavations for the purpose of determining the history of the temple, I shall confine myself to the large trench in front of the ziggurat and to the results obtained in the excavation of the ziggurat itself, premising that there are still many matters which need to be worked out more fully, and which can be only tentatively given until the full re-

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

21

suits of Mr. Haynes' work are at hand, and have been compared carefully with the result of the work of the first two years. Before excavations began the ziggurat was an almost conical hill, known to the Arabs as Bint-el-Amir (Daughter of the Prince). Around this, on all sides, was a plateau, seamed here and there with very deep gulleys. The general height of this was 14 metres (see for general appearance of mounds P L A T E IV). On this plateau we found, first of all, a surface layer of to feet of earth. In this, close to the ziggurat, on the southeast side, were poor walls of mud brick, remains of a number of rooms or huts of a late period. To the northwest of the ziggurat we found two or three Jewish bowls, such as were also found in great numbers in a Jewish village on another hill, in which latter case coins of the Kufic period gave us as a date the v n century after Christ. Here and there on the plateau of the ziggurat were coffins and tombs, which are to be ascribed variously to the Parthian, Sassanian and Arabic periods, although no remains were found which would enable us to date any of them with precision. The remains in this upper layer of earth point to a time when the temple was no longer a temple, but when the plateau and Bint-el-Amir were merely a tel, the latter affording protection, and a side hill for the building of huts, and the former a suitable place for burial, according to the ideas of the people. In thia stratum, very little below the surface, was a layer of fine white ashes, pretty evenly distributed over the surface of the plateau, evidence apparently of the use of the hill by alkali burners. Below this later stratum, or these later strata of earth, we came to a series of constructions which belonged together, constituting one whole (Fig. 3). Walls of unbaked brick stood to the height of 14J to 15J feet. To the southeast, northeast and southwest of the ziggurat were rooms or houses; to the northwest and north were very fine series of corridors. The whole, as indicated in the plan (PL. I l l ) was bounded by a vast retaining wall. On the southeast side the rooni3 or houses were contained within an inner wall, which was relieved by two singular solid towers of a conical shape. Outside of this wall, and between it and the great retaining wall, was a huge corridor. Through the inner wall was no door, but passage to the ziggurat was obtained by a circuitous

22

JOHN P.

PETERS.

route around to the south and southwest.

T h e walls of t h e

rooms and corridors of this series were in almost all cases

finely

stuccoed w i t h a plaster of mud and straw, smoothly laid on, and many of t h e m had been tinted, but always seemingly in solid colors. T h e accompanying sketches and plans by M r . Joseph A . M e y e r , J r . , w h o has been w i t h M r . H a y n e s for the last year as draughtsman, will give some idea of the appearance and character of the rooms to the southeast and east of the ziggurat.

FIG.

3.—ROOMS

S.

E.

OF

ZIGGURAT

or

E X C A V A T E D AT BEGINNING

LATE

BABYLONIAN

OF SKCOND

A l l of

the

PKRIOD.

YKAK.

rooms on this level w e r e occupied during t w o or three successive periods, as is shown by the w a l l i n g up of doors on lower levels, and the opening of others on higher levels, the building

of

threshold upon threshold, etc.

of

T h e accompanying sketches

M r . M e y e r g i v e a specimen of a house occupied m three or possibly in four successive periods, as shown b y the doors.

I can-

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR

23

not do better than to repeat Mr. Haynes' words with regard to these sketches: " Fig. 4 gives a sketch in perspective of the street shown in the accompanying plan (Fig. 5). It looks towards the southsouthwest, and shows its continuation along the face of the south-

FIG.

4 — A S.

E.

STREET OF T H E

IN T H E

TEMPLE

ZIGGURAT

(July

ENCLOSURE, 5,

1894).

eastern buttress of the ziggurat, and under the steps on the opposite side of the great trench. In the middle of the unpaved street is a well-made gutter of burned bricks. The masonry of combined crude and burned bricks, in the left hand middle distance, shows a stairway descending from the lilled-up street of what seems to be a well-defined period in the occupancy of these houses, and the continued use of the street, to the lower room of a house, that was continuously occupied and kept free from accumulating earth long after the street and the neighboring houses had become filled to a higher level with earth and debris, even after the doors of other houses had been raised to enter the street over a low threshold, sometimes of burned, often of crude bricks. . . . The walls of these houses clearly show three distinct periods in their occupancy. After the first occupation, during which time the street and many of the houses were filling with earth,

24

JOHN

P.

PETERS.

the walls of crude bricks were at least twice raised to a higher altitude, and twice were the doors carried upward to a corresponding height. The house in the left of the accompanying sketch (Fig. 4) shows three doors marking the three clearlydefined periods in the history of these houses. The lower door, with the segmental arch, belongs to the first period; the second door was closed by a mass of crude brick, which projected beyond the face of the walls; and the opening above this projection is the ruined door of the third and last discernible period. " F i g u r e 4 also shows toward the right hand sketches of two fireplaces. . . Figure 6 gives the cross-section and groundplan of the street at the point of the door shown in the left hand of the foregoing sketch. It also shows cross-section and elevation of the doors of the first and second periods, belonging to the above houses, and shown in the sketch and in the section and plan of the street given in Figs. 4 and 5. " F i g u r e 7 is a sketch of the domestic pottery taken from this series of rooms. . . The large vase in the centre of the group was perforated—probably to allow the escape of water. It was sunk below the floor of earth in the northern corner of the room numbered 121 on the accompanying plan (Fig. 5). Around

Scale, .021 in. = 1 ft.

the jar's mouth was a bit of bitumen cement, apparently designed to convey water into the jar, which would seem to have served as

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

25

26

JOHN P.

PETERS.

a drain, although a more unsanitary method of plumbing could scarcely he devised. Drainage was generally effected by a sluice under the threshold of the door into the street, or by a small drain through the wall of the house, also into the street, the sloping floors facilitating the fall of the water into the drain."

'FRSN HOW. ®F- THE TEMFIF /1

FIG. 7.—DOMESTIC

POTTERY

FROM

HOUSES S.

E.

OF T H E T E M P L E

OF

BEL.

Figure 8 exhibits in detail the curious and interesting door socket and threshold of burned brick seen at G on Fig. 6.

PIG. 8.—DOOR-SOCKET

OF B U R N E D

BRICK

WITH

P A R T OF T H E

THRESHOLD.

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT

NIPPUR.

27

Figure 9 exhibits the drains and ventilators found in the walls of houses of the first of the three periods here represented.

mm - j ^ X .

. miu «"A pipf.

i

HTOTION

^

n r T

]

^ECTIO/V

N« I

^FCRI«N

N!2

7 i

EE lVATION tl!2 FIG. 9.—VENTILATORS THE FIRST

OR D R A I N S

PERIOD S. E.

F O U N D IN S T R E E T

OF T H E Z I Q G U R A T .

WALLS

(Scale,

OF H O U S E

.445 in. =

OF

1 ft.)

It was evident that these rooms or houses were occupied during a considerable period of time. Some pottery and terracotta figurines of Greek work (PL. V, 1) show that a portion of that time at least was in the Seleucidan period, but there are no remains which enable us to fix either a terminus a quo or ad quern for these buildings. I am inclined to ascribe their origin to the later Babylonian empire, partly on the ground of general similarity to other structures of that period found in Babylonia. The cruciform shape of the ziggurat is, to be sure, unlike anything else which has been discovered, but the relation of these rooms or houses to the ziggurat is in general the same as that of the rooms or houses unearthed by Hormuzd Rassam atBorsippa to the ziggurat of the temple of ifebo built by Nebuchadrezzar. A s already stated, these buildings were removed in a line between the central portion of the ziggurat and the great wall to

28

JOHN

P.

PETERS.

the southeast, daring the second year of my excavations, and a large trench was conducted at that point through the successive •strata underlying these buildings. This trench was very much enlarged by Mr. Haynes and carried still deeper. The general results of the examination of the successive strata at this point were as follows: The houses or rooms described above had their foundations 18 feet below the surface, and rested upon earth, well packed together, 3 feet in depth. This again rested upon a mass of rubble and debris containing no walls, but great quantities of bricks and fragments of bricks with inscriptions of Meli-Shiha, and also bricks with green glazed surfaces. The same stratum was found at the same depth on the northwest side, on the southwest, and on the northeast—that is, on all four sides of the ziggurat. This would seem to indicate that there had been a very thorough demolition of some former structures before that restoration of the temple which gave the ziggurat its cruciform shape and surrounded it with the buildings of unbaked brick which have been described above. It would also seem that the last important builder before the reconstruction, which I have referred to the late Babylonian empire, was Meli-Shiha. This Meli-Shiha, who, as we shall find shortly, played a very important part in the reconstruction of the temple, was identified by Prof. Hilprecht, in his Old Babylonian Inscriptions (p. 55), with Meli-Shikhu, a king of Babylon of the Cosssean dynasty, who ruled 1171-1157 B. C. Now, however, on the basis of further examination of more inscriptions, Prof. Hilprecht reads the same not Meli-Shiha, but Ashurbanipal, thus changing the date to 669—626 B. C. The m iss of debris and rubble, as stated, was about 4 feet in thickness. Below this, to the southeast of the ziggurat, and extending as far as the inner wall with the round towers, indicated on the plans, there existed, apparently as late as Meli-Shiha's time, an open court paved in brick. Various fragments of pavements were found in different parts of this space, and at one place three successive pavements occurred within a space of five feet. Inscribed bricks found in some fragments of pavement show that one of these was the work of Ur-Nmib, king of Isin, perhaps about 2600 B. C. The other pavements consisted of uninscribed

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

29

bricks, and it is therefore impossible to assign to them a certain date until further excavations at the north and south shall discover buildings connected with this platform which may enable us to do so. During this whole period the line which is now marked by the inner wall on the plans seems to have been the boundary of this court toward the southeast, and the two conical solid towers which were exposed by my excavations as forming part of the last great reconstruction of the temple, appear to have been in existence at the same place during this whole period. These great towers I am inclined to compare with the columns called Jachin and Boaz before the temple of Yahweh, at Jerusalem, with the similar columns before the temple at Hierapolis, and before Phoenician temples. Similar columns also existed in temples in Arabia, and the towers found by Bent in his excavations in Mashonaland seem to be of the same character. I suppose these towers, therefore, to have been sacred pillars representing the principle of life—gigantic, conventionalized phalli. They do not differ in shape or position from the columns used in the temples mentioned, but only in material and size. A t about 30 feet below the surface of the plateau of the temple, to the southeast of the ziggurat, we came upon a pavement of Ur-Gur, about 2800 B. C. This was of crude brick eight feet thick at its thickest point. It constituted an enormous platform or terrace, on the northeastern edge of which stood the ziggurat, while the southeastern part, as far as the towers, was an open court. This court was flanked, at least on part of its northeastern side, by buildings, but its further dimensions we do not know. Immediately below this was found at one point another pavement of bricks of unusual size, 18 and 20 inches square, with a thickness of three inches. This pavement was identified by Mr. Haynes as a pavement of the time of Naram-Sin, king of Akkade, 3750 B. c., or of Sargon, his father, 3800 B. C., by comparison of inscribed bricks of these same monarchs found elsewhere. Immediately to the northwest of the temple is a plain, bounded on the northern side by a wall line, which seems to be the outer wall of the city at that point. I conducted excavations here in the second year of my work, and later Mr. Haynes did the same with

so

JOHN P.

PETERS.

greater success. This wall was found to consist in its lower part of unbaked bricks, stamped with the stamp of Naram-Sin, king of Akkade, the stamped face turned down. Immediately above this, with no intervening work, was found a wall of Ur-Gur. The bricks of Naram-Sin were of singularly large size, of wellmixed clay, tempered with chopped straw, carefully moulded and thoroughly dried, so as to attain an unusual hardness and firmness of texture. The bricks of Ur-Gur are of almost equal excellence, but of small size, and of a shape much resembling our ordinary modern bricks. So characteristic are the bricks of UrGur that it is generally possible to determine a structure of his without inscriptions. The bricks of Naram-Sin, and of Sargon, his father, seem to be equally characteristic in a quite opposite direction; and it seems, therefore, safe to assign this fragment of a platform to one of these kings, especially since their work was immediately below that of Ur-Gur in the outer city wall. It should be added that the platform of Naram-Sin or Sargon was not coterminous with that of Ur-Gur above it, for at various points in the great trench and elsewhere on the plateau I discovered walls of unbaked brick, among which were three door sockets of Sargon, king of Akkade—two of them apparently in place and one inverted. I also found at this level a clay brick-stamp of the same king. There were also found here quantities of vases and vase fragments, chiefly in marble, of a new king, " Alu-Sharshid, king of the city," who has been identified by Prof. Hilprecht as of the same dynasty, and approximately, therefore, of the same time as Sargon and Naram-Sin. Other fragments of vases of this latter king, it should be added, were found both by Mr. Haynes and myself above the level of the Ur-Gur platform, some of them containing a second inscription of that Gande mentioned above, whom Prof. Hilprecht identifies with Gandash, founder of the Cosssean dynasty, and who, therefore, ruled about 1450 B. c. It would seem that the stratum of the dynasty of Akkade, containing remains of the three kings, Sargon, NaramSin and Alu-Sharshid, lay in general immediately below that of the dynasty of Ur, represented by Ur-Gur as its great builder. The depth of the bottom of the Ur-Gur stratum was nine metres, of that of the Sargon stratum about 11.5 to 12 metres. Mr.

UNIVERSITE

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

31

Haynes continued the trench to a point some 16 feet below the bottom of the Ur-Gur platform, or about 54 feet beneath the surface, and sank a shaft to the depth of 68J feet, at which point he says that he reached virgin soil, although water was reached at a slightly higher level. My excavations were carried down systematically to a point 11.5 metres below the surface, and by a shaft I descended to the depth of 14 metres, or 46 feet. A t the depth of 12.95 metres I found a large jar, the same as those found through all periods down into the post-Babylonian. Below the depth of about 40 feet nothing was found by which dates could be fixed. It will be seen from the above figures that, according to the dates ordinarily accepted by Assyriologists, the upper 40 feet of accumulations in this great trench represent a period of not less than 4500 years ; how long a period was represented by the "28| feet below this level we have as yet no way of determining. Figure 10 gives a view from a sketch by Mr. Meyer of the condition of Mr. Ilaynes' excavations at the end of the great trench toward the ziggurat, and including the eastern corner of the ziggurat, on Sept. 10th, 1894, after the great southeastern wing or buttress and the late ziggurat wall built over that of Ur-Gur had been removed. I will quote Mr. Ilaynes' explanation of this drawing : " A is Ur-Gur's burned brick wall, forming the façade of the lowest stage of the ziggurat. B shows a small part of the causeway (of Ur-Gur). C and I) are wells sunk to the water-level. E is the altar (this will be described later; its upper surface was nine feet below the level of the Ur-Gur platform). F is a curb of peculiar archaic-looking brick, seven courses high. It seems to mark a sacred enclosure, possibly extending around the earlier temple. It has been traced through the tunnels SS to the southwestern side of the great southeastern projection ol the cruciform temple of later Babylonian times. Its limits have not been found. (The same wall was found by me under the ziggurat, at its western corner.) G is a section of the pavement on which the bitumen construction was laid to protect the foundations of Ur-Gur's ziggurat from falling rain. This pavement, from front to back, is ten feet wide.

32

JOHN P.

PETERS.

II.—In a tunnel underneath this pavement at II is shown a section of the very ancient wall recently discovered. This wall has about the same batter as the later wall of Ur-Gur above it. Both

walls are in the same plane. Hence it can be literally said that Ur-Gur built upon older foundations, although a thick platform lies between the two walls.

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

33

K.—The pavement K K lies eight feet below the foundations of Ur-Gur's ziggurat. I t extends at least forty feet toward the northeast. Its limits toward the southeast are a matter of conjecture. The bricks in this pavement are about the same size and mould as the bricks of Sargon and Naram-Sin already found. Ur-Gur laid his great platform on the level of this pavement. M.—The pavement M apparently belongs to the construction whose wall is marked OOO. 1ST.—N is a very old wall, which evidently antedates Ur-Gur, since it interrupted his platform, and lies wholly within the platform itself. OP represents a tunnel following the ruined wall 0 0 0 toward the northeast, a distance of eighty feet and three inches, to the eastern corner of the building. The foundations of this building and the temple of Bel are on the same level. A t least the southwestern end of this building seemed to have been filled solid with crude bricks of the Ur-Gur size, form, color and texture. Could it have been a temple to Beltis, repaired by Ur-Gur?" Figure 2 represents excavations in the same trench three months earlier, and gives a good view of the causeway of UrGur, which was just visible on the extreme left of Figure 10. The walls of this causeway are each four feet thick, of burned bricks, most of them stamped with the name of Ur-Gur, laid in bitumen. These walls are nine feet apart, and the intervening space is filled with libben or unbaked brick. Figure 11 gives a more detailed view of the archaic curb marked O O

FIG. 1 1 . — A N C I E N T

CURB

FOBTY

FEET

BELOW THE

SURFACE..

F on Fig. 10, and Fig. 12 represents one of the plano-convex "bricks of this curb. The foundations of this wall Mr. Hayne

34

JOHN P.

PETERS.

reports as being 13J feet below the bottom of the Ur-Gur wall, which would make it identical in level with the wall of precisely similar construction found by me under the western corner of the

FIG. 1 2 . — R U D E

BURNED

BRICK

FROM T H E A B O V E A R C H A I C

CURB.

ziggurat, and render still more probable Mr. Haynes' proposition that it enclosed a sacred area. The remains of this wall were 18 inches high, and the bricks were laid in mud in courses alternately lengthwise and crosswise. It seems to be older than the time of Sargon, as it is below his level. Before the ziggurat, beneath the Ur-Gur level, in the stratum elsewhere assigned to the Sargon dynasty, was found an oven, in which was discovered a new baked clay tablet. This tablet has not yet been deciphered. Like some other tablets found at similar low levels, it is inscribed on one side only. Immediately above the Ur-Gur wall, in the same trench, were found a number of other tablets of most exquisite workmanship, which, like this, have not yet been deciphered. To the northeast of the ziggurat projection, was found a pottery drain in place. The accompanying Figures 13 and 14 give excellent illustrations of different forms

FIG. 13.—TERRACOTTA

DRAINS

FOUND

IN M O U N D

NO.

8.

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

35

of these drains, so characteristic of the ruin mounds of Babylonia from the time before Sargon up to the latest period. Sometimes they are composed of rings made for the purpose, sometimes jars are broken at top and bottom and fitted together, as in Fig. 13, No. 2. These drains are often 40 or 50 feet in depth. I have now given a survey of the strata unearthed in the great trench to the southeast of the ziggurat. More important results were obtained by the careful excavation oi the ziggurat itself. It was ascertained that the ziggurat which forms the core of the existing structure was the work of Ur-Gur. Toward the northeastern edge of his solid platform of unbaked brick, eight feet in thickness,' Ur-Gur erected a zisrO gurat in three stages. The lowest of these stages was 20J feet in height, the sides sloping upward at the rate of one in four. The second terrace set back 13J feet from the surface of the one below it. The height of this terrace at its slope I am unable to give, as also that of the terrace above. The lower terrace was faced with burned brick on the southeastern side looking toward °

PIG. 14.—TERRACOTTA D R A I N I n l e t FR0M M o u N D v m

>

AND LOWER SECTION o r THE SAME DRAIN.

the great open court. On all of the other sides there was a foundation of baked bricks four courses high and eight wide, above which the material used was unbaked brick covered with a plaster of fine clay mixed with chopped straw, which, being often renewed, preserved the crude bricks beneath as well as if they had been burned by fire. In the middle of each of these three sides was a conduit for the purpose of carrying away water from the upper surface of the ziggurat

36

JOHN

P.

F I G . 1 5 . — W A T E R CONDUIT ON T H E

PETERS.

S. W. FACE

OP T H E

ZIGGURAT.

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

37

(Figures 15 and 16). This conduit was built of baked brick, and bad an inner breadth of 3 J feet and a depth of 10\ feet. There was apparently a similar arrangement for carrying off the water

FIG.

16.—WATER CONDUIT ON

S. W. S I D E OF THE ZIGGURAT.

(.Scale, .078

in.=lft.)

in the second and third stages; but it was ruined beyond possibility of restoration. Indeed, both of these stages were so ruined by water that it was difficult to trace or to restore them. Around the base of the ziggurat, on all sides, was a plaster of bitumen, sloping outward from the ziggurat, with gutters to carry off the water. By this arrangement the apparently very perishable foundation of unburned brick was thoroughly protected from destruction, and unburned brick, protected like this, is, at least in the climate of Babylonia, one of the most imperishable materials of construction that can be found. The first important change in the form of the ziggurat was made by Kadashman-Turgu, 1257-1241 B. C. He built around the ziggurat of Ur-Gur on three sides, at the base, a casing wall of brick sixteen courses, or 4 J feet, in height, but preserved and utilized the conduits of Ur-Gur. His is the wall exhibited at the base in Figures 15 and 16. The next great reconstruction was undertaken by Meli-Shiha. Upon the casing wall of Kadashman-Turgu he erected at a slightly different angle, and somewhat set back from the other, a second wall. The conduits on the southwestern, northwestern and northeastern sides he filled up with bricks—many of them

38

PIG.

JOHN P.

17.—N.

E.

F A Q A D E OF T H E

PANELED LATER

WALL

"WALL

or

ABOVE.

PETERS.

ZIGGURAT SHOWING MKLI-SHIHA

AND

A

TIIE

BUTTRESSED

FRAGMENT

OF

OR THK

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

39

stamped with his name—and the upper part of the lower terrace he faced on these three sides with a paneled wall of brick (Fig. 17), giving to the ziggurat quite a different appearance from that which it had hitherto possessed, and enlarging its dimensions, so that when left by him it measured 170 feet by 125 in length and breadth, or very little less than the ziggurat of the temple of Sin at Ur. The reconstructions of Kadashman-Turgu and of MeliShiha seem to indicate a filling up of the surface immediately about the ziggurat by the washing down of mud from above. This process continued through the ages until, as we shall see, the greater part of the ziggurat was ultimately buried beneath the accumulations washed down from its own upper surfaces. Fig. 18, like the rest, from Mr. Meyer's drawings, gives a section

FIG.

1 8 . — SECTION OF N . E .

WALL

E. CORNER.

{Scale,

OF Z I G G U R A T , T H I R T Y F E E T FROM

.100 in. = 1 ft.)

view of the northeastern wall of the ziggurat after Meli-Shiha's restoration. Within is seen the original wall of Ur-Gur, the two lower courses of which continue outward and form the two lower courses of Kadashman-Turgu's wall also. In front of IJr-Gur'a wall, on this side, was a filling of crude brick three feet in thickness, and in front of this Kadashman-Turgu's wall of about the same thickness. Kadashman-Turgu's wall is 4 J feet high, rising, with the foundation of Ur-Gur beneath, to the six foot level.

40

JOHN

P.

PETERS.

Upon the wall of Kadashman-Turgu, for three feet, stands the foundation of Meli-Shiha's wall, at a different angle, as stated ; and above this, but setting back from it, the paneled wall described above rises still nine feet further. The drawing shows the peculiar curvature of the upper wall of Meli-Shiha. Figure 19 represents the elevation of the northeastern side of the northwestern façade, and shows the later structure erected on Meli-Shiha's wall. A represents the wall of KadashmanTurgu, B and C that of Meli-Shiha in its two parts. On this

FIG.

19.—ELEVATION

AND

PLAN

AT N . E. END.

OF THK N .

W.

F A Ç A D E OF T H E

ZIGGURAT,

(Scale, .042 in. = 1 ft.)

was built, at a later date, a wall of unbaked brick (D), of which three courses remain. The crude bricks of this wall are the characteristic bricks of the great reconstruction of the temple which gave the ziggurat its cruciform shape, and which covered the ground about the ziggurat on all sides with the rooms, houses and corridors shown in the plans. These bricks are large, almost square, of rather rough work—in many casés pieces of pottery being used to fasten the clay together in place of straw. The builder who erected this wall upon that of Meli-Shiha also added the wings or projections on all four sides of the ziggurat, and built over almost the entire ziggurat a new construction of unbaked brick, reducing at the same time the number of stages from three to two (Fig. 20). The rooms about the zigguratwere dovetailed into the new structure. At a later date the brick wall,

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

41

marked E on the plan, was built upon the remains of the wall of unbaked brick (D). This wall is of very late date, and composed, not of bricks made for the purpose, but of bricks taken from other constructions, so that the names of a large number of kings are found upon the bricks in this wall. Mr. Haynes suggests that at the time when this wall was built the ground about the ziggurat had been raised by the mud washed down from the surface to the point marked D, so that this wall was really built upon the surface

FIG.

20.—WALL

OF ZIGGURAT,

N. W.

SIDK, SHOWING LATEST ADDITIONS.

at that time. This seems quite probable in view of the relation to this wall of the late structures built above the level of the houses on the plateau. W e have thus rapidly surveyed the history of the ziggurat in its reconstructions; but it must be added that other kings did work upon both the ziggurat and the temple, besides those who are responsible for the great reconstructions. An examination of two of the corners of the ziggurat showed that at some time they had been removed almost down to the bottom and afterwards built up again. The bricks of both Ur-Gur and Meli-Shiha were

42

JOHN P.

PETERS.

originally laid in bitumen, but the bricks at the corners of the wall were laid in mud mortar (only those of Meli-Shiha and UrGur having bitumen adhering to them) thus giving evidence that these corners had been removed for some purpose and then built up again. Among the bricks of other kings found in the ziggurat were those of Bur-Sin, Ishme-Dagan and Kurigalzu, and among those found elsewhere in the temple were bricks of Bur-Sin of Isin and Esarhaddon of Assyria, showing that many kings of many ages had honored the temple of Bel at Nippur. Only Nebuchadrezzar ii of Babylon, the great builder of temples in other parts of Babylonia, is conspicuous by his absence. And now, to go back to the ziggurat of Ur-Gur, it will be asked, " W h a t was the object of the ziggurat?" " H o w was access had to its upper terraces ?" and " Where, if at all, were sacrifices offered in connection with i t ? " In answer to the first question, I would say that this particular temple of Bel, the Lord, whose proper name was En-Lil (Lord of the Storm), was itself known as E-Kur, or Mountain House. From a comparison of this ziggurat with others, in Assyria and Babylonia, with those described as existing in Southern Arabia, and with Jewish, Phoenician and Syrian temples, as we have them described in the Bible and other ancient sources, from a consideration of the traditions of mountain-worship existing among Semitic peoples, which we find so well illustrated in the Bible by the high places and also by the traditions regarding Sinai, Horeb and the like, and above all from a study of the description which Herodotos gives of the temple of Bel Merodach at Babylon, I have been led to suppose that ziggurats (the word seems to mean peak or high place) were nothing more than conventional mountains, and that the Deity was conceived of as inhabiting a Holy of Holies on the summit of these mountains, where he dwelt unseen, enshrined in darkness. Herodotus describes a chamber containing no image as existing on top of the ziggurat of Bel Merodach at Babylon. There were found at the top of the ziggurat at Nippur large numbers of bricks which seem to have belonged to some structure, although no walls could be found in place. It will be remembered also that on the summit of the ziggurat of the temple

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

48

of Nebo, at Borsippa, there are great masses of brick. In the case of the ziggurat at Nippur, I am inclined to think that these bricks represent a chamber or shrine which once stood upon the summit of the ziggurat, like that described by Herodotus as existing on the top of the temple of Bel Merodach at Babylon. In answer to the second question—" How were the upper stages of the ziggurat reached ?"—we found, as stated, a causeway running from a point at about the outer great wall of the temple, southeast of the ziggurat, up to the ziggurat. It was impossible to determine whether the ascent on this causeway was by steps or by an inclined plane, although I suppose the latter to be the more probable. Access was had from the lower terrace to the upper terraces apparently by a continuation of this same causeway. In answer to the question, " Where, if at all, were sacrifices offered in connection with the ziggurat?" I would say that beneath the platform of Ur-Gur, in front of the southeastern side of the ziggurat, between the causeway and the eastern corner, stood (as shown on Fig. 10) an altar. This altar was of unbaked brick some 13 feet in length. On this was a ring of bitumen seven inches in height, the surface within which was wholly covered with ashes; some of them bone ashes. To the southwest of this altar was a sort of bin or receptacle of crude brick, full of ashes to the depth of about a foot. This altar was apparently in use at the period of Sargon, although its foundations may have been more ancient. No altar was found at any other level, but I am inclined to reason from this altar to the position of the altar at all times, and suppose that sacrifices were offered at the foot of the ziggurat, on its southeastern side—-just as in the temple of Yahweh at Jerusalem sacrifices were offered upon an altar which stood outside of and beneath the elevated or holy place on its eastern side. The Holy Place and the Holy of Holies in the temple at Jerusalem seem to me to correspond to the ziggurat and the chambers upon it in the temple of Bel at Nippur, in the sense that the temple at Jerusalem was a development from a ziggurat temple, like that of Bel at Nippur. The notice of this altar which stood below the platform of UrGur leads me, finally, to note the constructions found beneath the

44

JOHN P.

PETERS.

Ur- Gur level. In front of this altar, at a distance of six feet from it, as already noted, ran a wall of most archaic construction, which also appeared under the western corner of the ziggurat, and which may at some time have been the boundary of an inner Holy Place—the court of the priests. In exploring at Mughair I thought that I found traces of a similar wall around the ziggurat of the temple of Sin at that place. Beneath the platform of Ur-Gur, under the eastern corner of the ziggurat, was found a construction which, from the bricks composing it and about it, Mr. Haynes thinks to have dated from the time of Naram Sin, or his father Sargon. The walls, of admirable construction, were standing to the height of 11 feet, and the platform of Ur-Gur's ziggurat rested immediately upon these walls. The building of which these walls formed a part proved on examination to be a solid tower 23 feet square, but no ziggurat—unless, indeed, this were the lower stage of a ziggurat of very small dimensions. No traces were found of any ziggurat earlier than the time of Ur-Gur, unless the tower above referred to was, as suggested, a ziggurat of very small dimensions. Mr. Haynes raises the question whether Ur-Gur was the first builder of ziggurats in Babylonia, and calls attention to the fact that the earliest ziggurats known, those of Nippur, Erech and Ur, are all his workmanship. That there was from a much earlier time a temple at Nippur on the same site as in the days of Ur-Gur, and with the same name, is shown by the inscriptions of Sargon and Alu-Sharshid. The only question is as to the form of that temple. Beneath the tower mentioned above Mr. Haynes found an arched drain, the arch of which he describes as " Koman," but no full description and no drawings or photographs of it have yet come to hand. He also found bricks laid in lime mortar. He also found at the same depth a great quantity of clay water-cocks (Fig. 21), which, as he points out, are identical with those manufactured and used as drinking-fountains in many parts of Turkey to-day. To get a drink one closes with the hand the small lower orifice, whereupon the water fills the cock, and one drinks with the mouth directly from the large upper orifice.

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

45

There were also found at this same level, and near to the square tower, some fragments of unbaked tablets, together with

FIG.

21.—WATER

TENT

FORTY-FIVE THE

FEET

BELOW

THE SURFACE S.

E.

OF

ZIGGURAT.

pieces of clay prepared or being prepared to receive inscriptions, thus proving that as early as the time of Sargon (3800 B. C.), clay tablets were used for purposes of writing. (It may be said that the earliest dated tablet yet found anywhere, which was found at Nippur, is a tablet of Dungi, the son of Ur-Gur.) Fifteen feet lower than the level of the drain, the water-cocks, and so forth, but in the great trench (1 on the plan), and not under the ziggurat, Mr. Ilaynes also discovered a mortar of burned clay, a stone vase perforated with several holes, and a gutter of burned clay, meant to serve as a gargoyle to conduct water from the roof. The fragments of pottery and the bricks found at this and even lower levels were practically identical with those found at much higher levels, showing a homogeneity of civilization and culture throughout. In other words, from the stratum of the late Babylonian constructions down to a stratum twenty feet and more below the stratum of Sargon, we have everywhere the remains of a high and practically homogeneous civilization. Now, in view of the mass of accumulations beneath the Sargon level, if, with practically all Assyriologists, we accept for Sargon the date 3800 B. C., we must suppose the earliest constructions on the site of the temple of Bel at Nippur to have been erected as early as 6000 B. C., and perhaps even earlier, and that civilization in Babylonia had been carried to this high state at that very early date. (On geological grounds I have argued that the foundation of Ur and Eridu was between

46

JOHN P.

PETERS.

6000 and 7000 B. c.) There is, according to the measurements given above, 20 feet of débris to be accounted for before the time of Sargon, and we can hardly assign for this a shorter period than 2000 years. After Sargon's time the dating of the strata and the rate of accumulation seem to be satisfactory until we reach the stratum of Meli-Shiha. I f he were Meli-Shikhu (1171-1157 B. C.), as at first supposed by Prof. Hilprecht, the general proportions of accumulation would be approximately correct; but if he be Ashurbanipal of Assyria (669-626 B. C.), as Prof. Hilprecht seems now to have proved him to be, then the accumulations between him and his immediate predecessors seem to have been abnormally slow, and those after his date as abnormally rapid. Whoever he was, he was one of the greatest of the builders of ïiippur and the most artistic. He used burned bricks and glazed bricks freely, and some of the brick constructions embedded in the later masses of unburned brick presumably date from him. After his time came a great catastrophe. Everything except the ziggurat was razed and rebuilt, and even that was totally changed in appearance and built over, if not rebuilt. Who it was who thus rebuilt the temple we do not know, although on general grounds I have assigned this work to the neo-Babylonian period. It certainly cannot be later than that ; and if Meli-Shiha were not Ashurbanipal, I might even have supposed it to be earlier. Whoever did the work was certainly a great builder, and the walls, terraces and the like of this reconstruction are really of astounding size and fine construction. Nebuchadrezzar is the only later monarch of whom I can think who would have been likely to have undertaken a work of such dimensions ; but as he used burned bricks stamped with his own name more freely than any monarch of whom we know, the absence of such bricks from this construction seems to be proof positive that he was not the builder. JOHN P .

St. Michael's Church, ISTew York, March 6, 1895.

PETERS.

JOURNAL

OF

ARCHEOLOGY.

J_J

L

J***-

VOL.

X.

PLATE

III.

i I I 13 •r , n r r U }>N»U{

¡•»HV^j HtA»| — •SI

JOURNAL

OF

ARCHAEOLOGY.

RAISED

MAP

VOL.

OF T H E

FIRST

FROM D R . H I L P R E C H T ' S R E P O R T

44

YEAR'S

EXCAVATIONS

AT

X.

PLATE

IV.

NIPPUR.

T H E BABYLONIAN E X P E D I T I O N , " V O L . I .

(The Roman Numbers indicate places where excavations were made ; the Arabic, the height of the mounds, in metres above present level of canal bed. About five metres should be added to obtain actual height above plain. III is the Temple-mound, Ekur ; VII is the wall, Nimit-Marduk.)

JOURNAL

II.

OF

EARLY

ARCHAEOLOGY.

BABYLONIAN

VOL.

X.

PLATE

AND L A T E GREEK TERRACOTTA FIGURINES F O U N D TEMPLE OF BEL.

V.

IN

U N I V E R S I T Y OF P E N N S Y L V A N I A E X C A V A T I O N S AT NIPPUR. II. T H E N I P P U R ARCH. [PLATE

XX.]

In an article on the temple of Bel at Nippur, published in this J O U R N A L , Vol. x, No. 1, I mentioned the discovery by Mr. Haynes of a " Roman " arch. This gave rise to an unfortunate misunderstanding, as though he had meant to ascribe the arch to the Roman period. In point of fact, Mr. Haynes claimed from the outset that he had discovered a true arch far antedating any hitherto discovered. A t the time when my article was written I was unable to give any details regarding this arch, but since then the Committee has received from Mr. Haynes blue prints of the arch and its surroundings, which prove it to be a true key-stone arch, pointed, and older than the time of Sargon of Agane (3800 B. c.). The accompanying drawings, made from these blue prints by Mr. James T. Dye, will demonstrate, I think, the complete accuracy of Mr. Haynes' claim that he has discovered a true arch, older by many hundred years than any hitherto known. It will be interesting to give the history of the discovery in his own words. Under date of Oct. 18th, 1894, he wrote as follows : " Underneath the spot where the greatest number of these terracotta water-vents were found [an illustration of these water-vents was given in the afore-mentioned number of the J O U R N A L , F I G . 21] we have to-day come upon a drain extending under the walls of the aforesaid building. The drain appears to be older than the building above, and to have fallen into disuse before the building was placed above it." A week later, Oct. 20th, he writes: " T h e drain reported in my last letter to have been found under the very ancient building 352

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

353

or edifice under the eastern corner of the ziggurat has been followed out, and at its outer or discharging orifice we have just found a section of an arch that may have originally covered the whole drain. This is a perfectly formed elliptical arch of one foot and eight inches span, and one foot one inch rise, with a total height of two feet four inches from the bottom to the top of the arch." And a month later, Nov. 24th, he writes that " the drain passes under the entire breadth of the edifice."

FIG.

41.—VIEW

OF A R C H FROM T H E

INSIDE.

PIG. 41 gives a view of the arch above described " from the inside, before its front was opened. Two drain tiles are dimly seen

854

JOHN P.

PETERS.

in the bottom of the arch." FIG. 42 gives " a more distant view" of the same arch. PLATE XX " shows the outward side of the arch. The arch here is forced out of shape. It would seem to have been done from the unequal pressure of the settling mass above itj when it was drenched, perhaps with percolating rain water, from above. Since the arch is laid in clay mortar the bricks would readily yield to unequal pressure, especially as these bricks

FIG.

42.—A

MOKE DISTANT

VIEW

OF

ARCH.

are convex on one side, while they are flat on the other side. You will observe one of the tiles (broken) in the bottom of the drain and a smaller tile in the top of the arch. I do not profess to know the meaning of these tiles. It is, of course, possible that the water-vents [which, as stated above, were found very near this arch and drain] served some purpose in connection with the tile in the top of the arch. The size of the tile admits of such

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT

NIPPUR.

355

possible use." Mr. Hay nos then calls attention to the proximity of the drain and arch to the altar, and suggests the possibility that the drain carried away the waste from the altar, while the small upper tile, to which was probably attached a water-vent, brought water for drinking and other purposes. The remaining illustrations (FIGS. 43, 44, 45, 46) show the position of this arch in relation to the surrounding and superincumbent structures. FIG. 43 " gives a front [southeast] view of the

Fio.

43.—VIKW

OF Z I G G T J K A T F R O M

SOUTHEAST.

ziggurat. It was taken from an opening in the great enclosing wall of the temple area in front of or southeast of the ziggurat itself. In the middle of the picture is the causeway, which may have been an approach to the higher stages of the ziggurat. It is composed of two parallel walls built of the burned bricks of UrGrur, many of which are stamped with the well-known eight-line inscription. The space between the two walls is filled with a regularly laid and solid mass of crude bricks, whose average

356

JOHN P.

PETERS.

dimensions are 9 by 6 by 3 inches. These bricks are of the same mould, and in color and texture are identical with the crude bricks composing the greater part of the huge mass of the ziggurat built by the mighty builder Ur-Gur. The stepped appearance of the two walls of the causeway is the result of cutting down the walls to make a level foundation for the façade or crust of the later cruciform construction [built against and upon the ziggurat]. As this construction was built up solid, the outer part or crust cannot be spoken of as a wall. It was under this crust, corresponding to the wall of a building, that the parallel walls of the causeway were cut down to provide against the ponderous settling of the mass above it. The tunnel under the entire length of the causeway proves the structure, as it now stands, to be homogeneous, and therefore the work of a single builder, who is the great builder of the ziggurat, which is now freshly exposed to view. " The original faces of the second and third stages of the ziggurat are respectively shown at D, D and E, E. B and C are central projections of the same stages. No such projections are to be found on any other side of the ziggurat. The design of these projections over the causeway is not evident." . . . . " "Whatever the purpose of this earliest causeway may have been, it seems to have suggested to the later generations the form that was adopted in the cruciform construction. At a higher level, and belonging to a later period than the causeway, were built from the middle of the four sides of the ziggurat, at right angles to its faces, four arms twenty feet wide and probably upwards of sixty feet in length. These arms were built of crude bricks, measuring 14 X 14 X 6 inches. " The cruciform construction of later times was a broadening of these arms on essentially the same foundations, thus making an immense elevated platform. It may readily be supposed that a smaller ziggurat . . . rose from the centre of this great cruciform structure as a platform. . . . This accounts for the large and high cone of crude bricks still rising far above the cruciform construction. "Whatever value one may assign to these suggestions, it is clear that the earlier causeway suggested the intermediate projections on the four sides of the ziggurat, and an enlargement of these produced the great cruciform construction."

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

857

In further elucidation of this illustration I may say that the shrine-like brick structure on the upper northeastern side of the ziggurat is not part of the ancient Babylonian temple, but a guard house erected by Mr. Haynes for his own protection at the excavations. The wall marked A is the face of the lowest stage of the ziggurat of Ur-Gur. On this side, and this side only, the lowest stage of the original was of burned brick, the remainder of the ziggurat being of crude brick, as stated in my recent article in the JOURNAL. T O the left of the causeway represented in this FIG. 43 was found a door-socket of trachytic rock with an inscription of Ur-Gur. The suggestion is that this door-socket1 originally stood on the causeway and was thrown down at the time when the later construction, described by Mr. Haynes, was built upon this causeway. In that case the causeway, as the approach to the ziggurat, was guarded by a gate. The form of the projections B and C, on the second and third stages, directly above the causeway, suggests some means of ascent to the summit, as by steps, at this point. The later reconstructions have, however, so modified the ziggurat at this point as to compel us to resort to conjecture. The cruciform structure which the ziggurat later assumed, whatever its origin, reminds one forcibly of the square cross, which I have found in Babylonia as early as the time of Gamil-Sin of Ur (2400 B. c.), and which symbolized the sun. This cross represents the two diameters of a circle, and may be used either with or without the circle about it. FIG. 44 " gives a good general view of the eastern corner of the ziggurat and the adjacent excavations." The wall of small baked bricks, broken into at the corner, is A of FIG. 43, the facing wall of the lowest stage of the ziggurat of Ur-Gur on the southeastern side. The brick wall visible on the northeastern side is of a later date, as explained in my late article in the JOURNAL. " T h e solid mass underlying the ziggurat of Ur-Gur, and included between the lines A—B, C—D, is a section of the platform of crude bricks (9 X 6 X 3), eight feet in thickness, which the first and the 1

A similar door-socket, found fifty or sixty years ago on or near the surface of the temple mound, is in the possession of a neighboring chief. I saw an impression of this stone in 1890, but was unable to purchase the original.

358

JOHN P.

PETERS.

greatest of the monumental kjngs [Ur-Gur] made, not only to serve as the foundation of his splendid ziggurat, but also to form the pavement of the entire temple enclosure, defined by the inner line of towers, of which the two bastions in front of the ziggurat are integral parts. Below the line C-D, but not extending so

F I G . 4 4 . — E A S T E R N CORNER OF THE ZIGGURAT.

far to the right as D, is the very ancient edifice descending eleven feet from the line C-D. There can be little doubt that it belongs to the time of, and is the work of some king of, the Sargon dynasty, or of an earlier king than even the very ancient Sargon." . . . . " In the line C-D [under the letter D] is seen a fragment of a pavement. The bricks of this pavement are the bricks of Sar-

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

359

gon and of his son Isararn-Sin. They are 15J X 15J X 3 inches in measurement." Under date of Kov. 24th, 1894, Mr. Haynes gives the following account of this pavement: " Underneath the crude brick platform on which the ziggurat was founded was a bit of pavement, consisting of two courses of burned bricks. The lower course of the pavement contained several stamped bricks of Naram-Sin, and at least three or four of Sargon's stamped bricks. The pavement contained bricks and half bricks of Sargon and his son, and may have been laid by the latter, or by some successor of him." That is to say, the bricks are evidently not old ones collected from other buildings or pavements by later kings and relaid at this poipt, but are found in their position as originally laid by Is aram-Sin. It should be noted that we always find 1ST aram-Sin in close association with his father, so close, indeed, that we might almost suppose that he was associated with him upon the throne; which association, if it existed, would well explain the use by Is aram-Sin of new bricks of his father along with his own. In confirmation of this date for this pavement are the additional facts that Mr. Haynes found at the eastern corner of the ancient building, immediately below the platform of Ur-Gur, a brickstamp of Sargon, and that while he found various objects with inscriptions of Sargon and Naram-Sin above this pavement he found nothing of either of these kings below it. My own discoveries of the remains of Sargon in so far confirm this view of the age of this pavement, that I found remains of Sargon and Alu-Sharshid immediately beneath the Ur-Gur remains. It must be added, however, that I also found at some distance away remains of Sargon at a depth of 7 | feet below this. So, also, in excavating the city wall to the northwest of the temple [xi in the general map of the mounds published in the J O U R N A L , P L . V] Mr. Haynes found crude bricks, 20 X 20 X 3 | inches, inscribed on the under surface with the name and titles of Naram-Sin, immediately beneath the familiar 9 x 6 x 3 bricks of Ur-Gur. But if this platform of two courses of baked brick were built by Naram-Sin, it is then clear that the ancient edifice, the f o u n d s tions of which are eleven feet below this platform, the altar, the

360

JOHN

P.

PETERS.

top of which is three feet below the platform, and above all the arched drain, the bottom of which is fifteen feet below the level of the platform, are older than the time of Naram-Sin. Mr. Haynes has been too modest to believe that he has made discoveries so much earlier than any heretofore made, or almost dreamed of, and in my article in the JOURNAL ( X , I) I followed too implicitly the example of his modesty. With the facts furnished by his latest letters and the accompanying photographs it is now clear that his later discoveries beneath the ziggurat are even much earlier than we had at first supposed. In my article on the Temple, I, following Mr. Haynes, ascribed the building beneath the eastern corner of the ziggurat to ITaramSin. It is clearly older, the pavement of the last-named king being flush with the summit of the remains of that building. The pavement, however, does not overlie this building, upon which, without anything intervening, rests the eight-foot thick platform of Ur-Gur, the foundation of his ziggurat. It was this lack of continuity of the pavement of iTaram-Sin, with the immediate superimposition of the work of Ur-Gur upon the tower, just as the work of Ur-Gur is superimposed upon that of ifaram-Sin in the external city wall, together with a partial resemblance in size and texture between the iJaram-Sin bricks of the city wall and the bricks of the ancient tower, which caused the mistake. Under date of Oct. 15, 1894, Mr. Haynes thus describes the ancient building beneath the eastern corner of the ziggurat: " A small and separate building . . . having an equal length and breadth of 23 feet, with a symmetrical and double re-entrant angle at its northern corner. It is built up solidly like a tower, and its exterior surface shows no trace of a door or opening of any kind. Its splendid walls, eleven feet high, were built of large crude bricks, each measuring one foot six and a half inches in length and breadth, and varying in thickness from 3J to 4 inches. [The iJaram-Sin bricks in the outer city wall measured 20 X 20 X 3J.] The bricks were made of tenacious clay, thoroughly mixed with finely cut straw and well kneaded. The batter of its wall averages f inches to the foot." FIG. 45 will give some idea of the relation to each other of this ancient tower (beneath which, it must be remembered, lies

UNIVERSITY

OF PENNSYLVANIA

EXCA VAT10NS AT NIPPUR.

361

the arched drain), the Ur-Gur ziggurat, the altar, and the archaic curb of brick described in the JOURNAL. The photograph from which the sketch was made was taken " from an elevated position nearly east of the corner." . . . . " A is the first stage of Ur-Gur's ziggurat. B is a pavement, about ten feet wide, on which was laid the sloping bed of bitumen to protect the foundations of the ziggurat from falling rain. The tunnels under this pavement discovered the lower archaic edifice that is still without a name. The curb of primitive bricks, seven courses high, supposed to

PIG. 4 J . — V I E W

OF A L T A R

AND

CURB.

bound the sacred enclosure around the altar, cuts off the view of the lower part of the tunnels. A wall of unexcavated earth is left underneath the curb to support it in place. C is the early altar lying under the eight-foot pavement of Ur-Gur, as did also the curb, which is still lower than the altar." The top of the altar, as already stated, was 3 feet below the bottom of Ur-Gur's platform. It was made of earth with a rim of bitumen around the edge on top. Its surface dimensions were

362

JOHN P.

PETERS.

13 feet by 8. It was well covered with ashes, some of which proved to be bone ashes, and a bin or receptacle, also of unburned clay, to the left (southwest) of the altar was half full of ashes. To the right of the tower is seen a part of the pavement of Naram-Sin. FIG. 46 will supplement FIG. 45 in explaining the relative position of the strata at the eastern coVner of the ziggurat, and especially of the arched drain. It is a view from the top of the

Fio.

46.—EXCAVATIONS

ABOUT A N D ABOVE THK

ARCH.

altar looking " down into the deep trench." " Gr is the curb. The early arch is directly under the curb, and being in deep shadow is scarcely discernible. The arch covers the mouth of an open drain seen at H. At I) is seen a bit of.pavement higher than the curb This pavement consists entirely of the burned bricks of Sargon and Naram-Sin. Directly upon this pavement is placed the great crude brick platform, eight feet thick, of Ur-Gur. Be-

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

363

low this pavement at D no bricks of Sargon or of his son XaramSin have been found. It should be noticed that this pavement . . . . is on the level of the top of the lower archaic edifice, underlying Ur-Gur's platform; that it is three feet higher than the ancient altar, and that it is eleven feet above the foundation of the lower edifice, and fifteen feet above the bottom of the early drain and arch." And now before summing up conclusions with reference to the arch and its date, it may be interesting to call attention to some of the objects found at or below the Xaram-Sin level, but above the level of the arch, near the ancient tower. " In a layer of light gray ashes, some four inches in depth, on the northeastern side of this building, and nearly on a level with the top of its walls, and underneath the ITr-Gur platform of crude bricks was found a fragment of an unbaked tablet," also " several lumps of kneaded clay, and among them an imperfect tablet prepared on one side only for an inscription." Besides the interest which this discovery has as showing the use of clay tablets at so early a date, it also exhibits the manufacture of tablets within the temple precincts, and in close proximity to, if not in connection with, the central shrine. It will be remembered that a pottery furnace containing a newly-baked tablet of a similar early date was found in front of the ziggurat to the southeast. More recently Mr. Haynes has discovered a deposit of unbaked tablets, apparently of the Cossaean period, in a room close to the western corner of the ziggurat. I found a number of beautiful baked clay tablets, unlike anything else which I have ever seen, quite close to the ziggurat to the southeast, but above the Ur-Gur level. All of which suggests to my mind a connection between the Temple, and particularly the ziggurat, and the manufacture of the tablets, especially in the earliest period, when we may suppose that writing was more rare, and hence more sacred. It is noticeable that almost every inscribed stone found at Nippur has been found in the Temple, and the very few fragments found elsewhere were manifestly not in their original position. " Several fragments of lime mortar have also been found in the debris near the walls of the above-mentioned building, and at a

364

JOHN

P.

PETERS.

depth of several feet below Ur-Gur's platform." The discovery of lime mortar would appear from this note of Mr. Haynes to have antedated Naram-Sin. The earliest use of bitumen for mortar which I remember to have observed at Nippur postdates this use of lime mortar, occurring in constructions of the time of TJr-Grur. As stated above, the bricks of the drain and arch were laid in mud. Considering the effect of running water on such mortar, one is almost inclined to argue that neither lime nor bitumen were known in Babylonia at the time of the construction of the arch. At the beginning of October, 1894, Mr. Haynes wrote : " On the southeastern side of this ancient edifice, nine feet below the bottom of Ur-Gur's platform, two terracotta water vents were found." Toward the close of November he writes: " On the southeastern side of the above-mentioned building [the archaic tower], and on a level with its foundations, have been discovered ten basketfuls of the archaic water vents." . . . . " All of these have been found within ten feet from the above-mentioned building . . . . and on the sides nearest to the altar." Attention has already been called to the possible relation of the archaic drain to the altar, and of the water vents to both. It is worthy of note that the necessity of holding and controlling water was one of the fertile causes in the early development of the art of baking and shaping clay in Babylonia. Among the apparently most ancient " finds " made by Mr. Haynes at Nippur was a terracotta fountain found in the bed of the Nil canal, which divided the city of Nippur into two parts. Under date of August 13th, 1893, he writes: " B y means of a trench 87 feet long, with an average depth of 21 feet, we have at length found the ancient bed, and northeastern, or left bank of the Shatt-enNil at the narrowest point of the main canal, opposite to the hill marked iv on the general map of Nippur [viz., PLATE I V accompanying my late article in the JOURNAL]. A t the depth of 20} feet below the surface, in the middle of the stream, and at the point where the accumulations above it were least, the bed of the canal was found." " In the debris accumulated above the bed of the stream, and seventeen feet below the surface, we found three fragments of an

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

365

ancient terracotta foundation of unique design, with interesting figures in high relief. One fragment, seven by ten inches, represents a priest clad in richly embroidered robes, and standing on the backs (shoulders) of two winged camels, I think possibly winged horses. Jets of water poured through the upturned heads of the animals. From the curvature of the fragments I judged the fountain to have been more than two feet in diameter, and there must have issued from it at least sixteen jets of water. To me these fragments are interesting from two points of view— first, as proving the existence of fountains at Nippur; second, as an example of somewhat archaic art, in which the perspective is bad and the species of the animals not easily distinguished, while the decorations and robes of the headless priest reveal the artist in a work of true merit." This fountain, together with the water-vents of terracotta and the arched drain with especially constructed tiles certainly show the importance of water works in the early art and architecture of Nippur. Mr. Haynes ventures the suggestion that the first use of baked bricks was due to the necessity of constructing drains and waterways capable of resisting the action of water. However this may be, the earliest arch yet found in Babylonia, or, indeed, anywhere, like the earliest arch found in Rome, the arch of the Cloaca under the Circus Maximus, was the arch of a drain or water-way. As has been already pointed out, this arch antedates by a considerable period the time of Naram-Sin (3750 B. C.), since it lies beneath structures which were themselves older than his era. It cannot apparently be ascribed to a period later than 5000 B: C., if the date of 3750 B. C. for Naram-Sin be correct. A more precise date I do not as yet venture to propose, as the strata below the L'r-Gur platform has not yet been explored over a sufficiently large area. Below the bottom of this arch also there are from twelve to fifteen feet of debris which are practically unexplored. As will be evident from the above descriptions of the arch and the position in which it was found Mr. Haynes has discovered a true arch of an almost incredible antiquity. After this article was already in type a letter from him under date of April 27th, 1895, announced the discovery of another arch, this time of crude

366

JOHN P.

PETERS.

brick, in hill x, a part of the city separated from the temple by the M l canal. He describes this arch as "pointed," by which I understand that he means sharp pointed, like the Gothic, and not blunt, or round pointed like the one described above. Mr. Haynes conjectured that this latter vault or arch might be older than 2000 B. G., and from the objects which he reports as found with it and about it his conjecture is confirmed, since these objects date from about 2500 B. C. We have, then, two arches from Nippur, one from about 5000 B. C., and the other from about 2500 B. C. The construction of the former of these arches shows us that at that very early period the principle of the arch was already thoroughly understood in Babylonia and that the arch already had a story behind it. To the best of my knowledge no other examples of the true arch have been found in Babylonia earlier than the Parthian or Sassanian period. This is due partly to the fact that so little excavating has been done among the ruins of that region, and partly to the fact that the upper portion of constructions of all sorts is the part which has almost always fallen completely into ruins. In Assyria, however, Layard found a vaulted room and more than one arch in the ruins of Nnnroud. H e reports these as true arches and says of one of them in his Nineveh and its Remains, Chap, x i : " The arch was constructed upon the well-known principle of vaulted roofs— the bricks being placed sideways, one against the other, and having been probably sustained by a frame-work until the vault was completed." A t Khorsabad, Place discovered several arched drains, pointed, elliptical and round, but in these the bricks or stones were laid at an angle, each course having a support in the course before it, so that no frame was required in the construction, a method of building arches employed to this day in the Turkish empire. In Egypt it is possible to trace somewhat more fully the development of the arch, but there also great lacunae are yet to be filled. The principle of the arch, support by thrust, seems to be recognized in the pyramid of Cheops, where the roof of one of the chambers, having an enormous weight to uphold, is formed by two stones resting against one another at an angle. The third pyramid, of the fourth dynasty, advances a step further. In this

UNIVERSITY

OF PENNSYLVANIA

EXCAVATIONS

AT NIPPUR.

367

the roof of one of the tomb chambers is formed of blocks of granite resting against one another at an angle, as in the pyramid of Cheops, but it is unlike the latter in that these blocks are hollowed out on the under side, thus giving an effect something like the English Gothic. A tomb at Abydos, of the sixth dynasty, described by Mariette, presents us with a keystone arch of elliptical shape in which the key and the two bases are of stone, while the intervening portions are of unbaked bricks, leaving large interstices to be filled in with mud and small stones. Two very ancient tombs at Sakkarah, the precise date of which is uncertain, exhibit the arch completely developed (Maspero, Mission archéologique française au Caire, 1,195), and by the time of the thirteenth dynasty elliptical and round arches become quite common in tombs. While it is probable that in this as in other matters the civilizations of Egypt and Babylonia were parallel and not dependent, so far as our present information goes, the arch was known in the latter country earlier than in the former, the Nippur arch, discovered by Mr. Haynes, antedating the earliest true arch yet found in Egypt by more than a thousand years. The earliest arch yet discovered outside of those countries, namely the Cloaca under the Circus Maximus at Rome, is nearly contemporary with the arches found by Layard at Nimroud, and more than four thousand years later than the earliest arch discovered by Haynes at Nippur. In conclusion, although it has no bearing upon the subject of this arch, or its date, I will take this opportunity to correct what now appears to be an erroneous statement in my article in the J O U R N A L on the Temple of Bel, on the basis of fresh information from the field. As I stated then, we are not yet in a condition to reach final results in many points, and, as all know, theories are apt to be overturned by new facts, even when we think we have them well established. I suggested that the two towers on the inner wall of the temple enclosure were pillars of the same nature as Jachin and Boaz in the Jewish temple, conventionalized phallic symbols. Mr. Haynes appears at length to have established the fact that they were bastions on a line of fortification •enclosing the temple court, erected by Ur-Gur, and rebuilt or built upon by others at a later date. His final proof, which

368

JOHN P.

PETERS.

seems convincing as to the intention of the towers, at their last reconstruction at least, is the discovery of a parapet, something which strangely enough I failed to find. It was the fact of the cone-like shape of these towers, precisely like gigantic phalli of a type very common at Nippur, in conjunction with their position, which reminded me of that of the solid cone-shaped structures found by Bent in Mashonaland, and the use of the two columns in Syrian, Phoenician and South African temples, as well as at Jerusalem, which led me in reconstructing the temple to form such a theory regarding them. It ought to be added that small phallic symbols are very common in the Temple at ISTippur. Some of these represent the male organ in the most completely naturalistic fashion, and from these to the inscribed nail-shaped objects, found in such large numbers at Tello, we have been able to form a complete and unmistakable series. These phalli were for the most part scattered promiscuously through the debris at all levels from the surface downward. Once only we found them in unmistakable connection with a wall, thrust into the bitumen mortar between the bricks, or lying at the bottom of the wall in a position which suggested that they had once been thrust into it. It will be remembered that at W a r k a Loftus found a wall constructed entirely of these cones, arranged in patterns. I do not remember any report from M. de Sarzec with reference to the use made of these cones at Tello, but in examining his excavations I saw a wall from which his workmen said that he had obtained a very large number of nail-headed, inscribed cones, where the cones were built into the wall without pattern or order in the bitumen between the bricks. There were certainly hundreds of these cones in the wall at the time that I saw it. W h a t was the meaning of this use of the cones I do not know, but that the cones were conventionalized forms of the phallus was clearly established by the series collected at Nippur. In view of the ubiquity of phallus symbols in Babylonian ruins, and their varying sources, I trust that I may be pardoned for my mistake in regard to the cone-like towers or bastions. JOHN P . PETERS. St. Michael's Church, New York, J u n e 15, 1895.

AMERICAN JOURNAL Vol. X.

OF

ARCHEOLOGY.

OCTOBER-DECEMBER, 1895.

UNIVERSITY

OF P E N N S Y L V A N I A

No. 4.

EXPEDITION

TO

BABYLONIA. III.

T H E C O U R T OF C O L U M N S A T N I P P U R . [PLATE

XXI.]

In a former article in the A M E R I C A N J O U R N A L OF A R C H E O L O G Y , Jan.-March, 1 8 9 5 , pp. 1 3 - 4 7 , I described at some length the excavations of the temple of Bel at Nippur. The site of the temple occupies but a small portion of the ruin mounds at that place, and far the larger part of our finds of inscriptions were excavated in other portions of the ruins. One large cache of fine baked tablets of the Cossaean dynasty was discovered in connection with a large building of most interesting character on the southwestern side of the Shatt-en-Nil, directly opposite the temple, in that part of the mounds marked I in the plate accompanying m y last article, and also in the plan of levels ( P L A T E XXI). In the first year of our excavations our camp was pitched on the highest point of the mounds on that side of the old canal bed, marked 24 metres on the plan of levels, near the figure I on that plan (FIG. 48). There was some delay in commencing excavations because, not having filed a topographical plan at the time of application for a firman, according to the law, it was agreed that after reaching Nippur we should not begin to excavate until such a plan had been prepared, and accepted by the Turkish government. 439

JOHN P.

440

PETERS.

During the few days while the plan was in preparation we were occupied in building our camp. For this purpose bricks were needed, and workmen were sent out to gather them wherever they could be found upon the surface of the mound. Some of the men engaged in this search found a brick structure just appearing above the earth in a gully beneath the camp to the northeast, and proceeded on their own responsibility to excavate the structure and remove the bricks. Some of the bricks which they brought in were inscribed. This led to an investigation of